ORNITOLOGIA NEOTROPICAL ______Volume 17 2006 No. 3 ______

ORNITOLOGIA NEOTROPICAL 17: 321–331, 2006 © The Neotropical Ornithological Society

BREEDING BIOLOGY OF THE GREEN-FRONTED LANCEBILL (DORYFERA LUDOVICAE) IN EASTERN ECUADOR

Harold F. Greeney1,2, Robert C. Dobbs1, Galo I. C. Diaz1, Scott Kerr1, & Jill G. Hayhurst1

1Yanayacu Biological Station and Center for Creative Studies, Cosanga, Ecuador, c/o Foch 721 y Amazonas, Quito, Ecuador. Email: [email protected] 2Research Associate, Museo Ecuatoriano de Ciencias Naturales, Rumipamba 341 y Av. Shyris, Quito, Ecuador.

Resumen. – Reproducción del Pico de lanza mayor (Doryfera ludovicae) en el este del Ecuador. – Breves recuentos sobre la biología de reproducción del Pico de lanza mayor (Doryfera ludovicae) han sido publicados anteriormente, pero aún no han sido realizados estudios detallados. Se observaron 36 nidos del Pico lanza mayor entre Septiembre del 2001 y Marzo del 2002 en el este de Ecuador. Los huevos presenta- ron una forma elíptica, fueron de color blanco inmaculado, y con medidas de 14,9 x 9,6 mm. El promedio del tomaño de anidada fue de 1,92 y del éxito del nido fue de 42,1% (0,58 juveniles volantones por nido). El período de incubación fue de 20–21 días y el período de empollamiento fue de 29 a 30 días. Se encon- traron nidos a lo largo de los riachuelos exclusivamente, predominantemente en bosque maduro, desde los 0,8 hasta los 6 m sobre el nivel del suelo. Las hembras alimentaron a los polluelos que se encontraron en crecimiento medio, con una dieta basada en “canarias” (Ephemeroptera) y otros insectos pequeños. Abstract. – Brief accounts of the breeding biology of the Green-fronted Lancebill (Doryfera ludovicae) have been published, but detailed studies are lacking. We observed 36 nests of the Green-fronted Lancebill between September 2001 and March 2002 in eastern Ecuador. Eggs were elliptical, immaculate white, and measured on average 14.9 x 9.6 mm. Mean clutch size was 1.92 and nest success was 42.1% (0.58 young fledged per nest). The incubation period was 20–21 days and nestling period was 29–30 days. All nests were along streams, predominantly within mature forest, from 0.8 to 6 m above the ground. Adult females fed half-grown nestlings a diet that included mayflies (Ephemenoptera) and other small insects. Accepted 28 February 2006. Key words: Natural history, nesting seasonality, egg, nestling, nest placement, clutch size, Green-fronted Lancebill, Doryfera ludovicae.

INTRODUCTION fronted Lancebill (D. johannae), although for- merly placed in the subfamily Phaethornithi- The two Doryfera , Green- nae (e.g., Peters 1945), are currently thought fronted Lancebill (D. ludovicae) and Blue- to be part of the basal radiation of the Tro-

321 GREENEY ET AL.

FIG. 1. Typical cylindrical nest of Green-fronted Lancebill suspended by rim from a sheltered portion of a fallen log, November 2001, 2100 m a.s.l., Yanayacu Biological Station, Napo, Ecuador. Photo by H. F. Greeney. chilinae (see Schuchmann 1999, Stiles 1999). disjunctly, the tepui region of northern South The Green-fronted Lancebill ranges from America. The Blue-fronted Lancebill occurs Costa Rica to northwestern Bolivia, while the primarily at lower elevations (400–1600 m; Blue-fronted Lancebill is confined to the east- Stotz et al. 1996) than the Green-fronted Lan- ern Andes of Colombia, Ecuador, Peru, and, cebill (1200–1800 m; Stotz et al. 1996), with

322 GREEN-FRONTED LANCEBILL BREEDING BIOLOGY

FIG. 2. Adult female Green-fronted Lancebill incubating in nest partially supported by a small, sheltered rock ledge, December 2001, 2100 m a.s.l., Yanayacu Biological Station, Napo, Ecuador. Photo by H. F. Greeney. some overlap in the eastern Andes of Colom- habits, thus significantly expanding knowl- bia, Ecuador, and Peru (Hilty & Brown 1986, edge of its breeding biology. We also place Ridgely & Greenfield 2001). Both are this information within the wider context of confined to humid forests, are hover-feeding lancebill and life his- trapliners, and construct cup-shaped nests of tory. moss, plant fibers, and spider webs, which are suspended from hanging roots or vines (Stiles MATERIALS & METHODS 1999). In Ecuador, Green-fronted Lancebills are encountered most commonly along fast- We made observations at elevations ranging moving forest streams and are rarely seen out- from 2000 to 2200 m a.s.l. on the privately side of mature forest. owned reserve of the Bustamante family, Few details on the breeding biology of Hacienda San Isidro, and the Yanayacu Bio- Green-fronted Lancebill are available in the logical Station & Center for Creative Studies published literature, which is limited to brief (00°35’S, 77°53’W), Napo Province, north- accounts in field guides and short notes based eastern Ecuador. While logging and dairy on 1–2 nests. Here we present data on 36 farming pressure have caused recent defores- Green-fronted Lancebill nests in eastern tation, large tracts of intact forest remain on Ecuador and compare our observations with the preserve and in the adjacent Sumaco and previous descriptions of the species’ nesting Antisana reserves. All nest observations were

323 GREENEY ET AL.

TABLE 1. Range of dimensions (cm) of Green-fronted Lancebill nests in eastern Ecuador.

Nest stage Inside cup diameter Inside cup depth Outside nest diameter n (nests) Early incubation 3.5–4.0 3.0–4.0 8.0–9.0 4 Late nestling 3.5–4.5 x 5.0–7.0 3.0–4.5 6.5–10.0 x 8.5–11.5 5 made in a roughly 600 ha area of primary for- hatching of that egg. We measured nestling est bordered by the Cosanga River on the east period from the hatching of the first egg to and the Las Caucheras road to the west. This the fledging of the first young. We observed area is relatively flat by Andean standards, 51 eggs and measured 18 to the nearest 0.1 extends roughly 2.5 km west of the Cosanga mm. We measured or estimated nest height to River, and is dominated by tree genera such as the nearest 10 cm from the bottom of the Vismia (Clusiaceae), Solanum (Solanaceae), nest to the ground or stream, and measured Nectandra (Lauraceae), Miconia (Melastomata- nest dimensions to the nearest 0.5 cm at 10 ceae), Ficus (Moraceae), Bunchosia (Malphigi- nests. aceae), Myrica (Myricaceae), and Alchornea (Euphorbiaceae). The canopy is occasionally RESULTS disrupted by Chusquea (Poaceae) Cecropia (Moraceae), Tibouchina (Melastomataceae), Nest material, form, and location. All nests were Baccaris, and Vernonia (Asteraceae). The constructed entirely of dry mosses, liverworts, understory is dominated by Cyathea (Cya- and spider webs. There was no apparent lin- theaceae), Piper (Piperaceae), Cestrum (Sola- ing. All were attached to the substrate with a naceae), Siparuna (Monimiaceae), Monina tough lip consisting of a thick mat of spider (Polygalaceae), Weinmannia (Cunoniaceae), webs. Most were tall and cylindrical (Fig. 1) and various Rubiaceae. Rainfall ranges from with a small cup at the top and often with a 2500 to 3500 mm per year, with the driest poorly formed “tail” of material hanging period generally being from August to Janu- from the bottom. All nests were located in ary. Rainfall during the study averaged 123 dark, moist, sheltered situations immediately mm per month, with the heaviest amounts in adjacent to, or directly over, 2–5 m wide September and March. streams. With one exception, all were located From September 2001 to March 2002, we within primary forest and ≥ 1 km from observed 40 nesting attempts. As four nests human disturbance. Most nests (23 of 36) were reused for a second brood, this repre- were pendant, and hung free from the sub- sents a total of 36 individual nests. We deter- strate or rested lightly against it. Of these, 19 mined clutch size by observing the contents nests were attached to rock or clay banks or of the nest at least three days after the laying overhangs. Two nests were in similar situa- of the last egg and for a period of at least tions, but were attached to live moss hanging three days. Only nests with incubating adults over rock edges and were “reversed” in orien- were used to determine clutch size. We tation such that the nest faced the rock cliff marked eggs with a small black dot at one end rather than away. One nest was attached to a using an indelible marker to determine order small rootlet forming a U below a rock out- of laying and hatching. Nests were checked crop and one was attached to the inner lip of once per day. We measured incubation period a rain culvert running under a road. The 13 from the laying of the second egg to the remaining nests were in similar situations, but

324 GREEN-FRONTED LANCEBILL BREEDING BIOLOGY

TABLE 2. Mean (± SD) measurements (mm) of Green-fronted Lancebill eggs.

References Length Width n (eggs) Ruschi 1961 16 11 2 Snow & Gochfeld 1977 15.7 (± 0.1) 11.1 (± 0) 2 This study 14.9 (± 0.5) 9.6 (± 0.2) 18 rather than hanging free, the cup was partially which approximated the size of the finished supported by a small ledge of substrate (Fig. nest but lacked any well-formed cup. Two 2). Small rock ledges supported 12 nests, and days later the cup was well formed and the one nest was partially supported inside a cav- female was observed entering the nest and ity under a fallen log. Nests with partially sup- wiggling around inside the cup, apparently ported cups tended to have shorter cylinders using her body and tail to further shape the but longer, partially formed “tails.” Nest cup. Twelve days from the original observa- height ranged from 0.8 to 6 m. tion, the first egg was laid. These and other Nest dimensions show that nest cups were observations suggest a 12–16 day building round early in incubation or before the onset period. of incubation, but that cups and nest cylinders At all of the nests except for those with stretched as incubation continued, and espe- running water directly below, we found dry cially during the nestling period, resulting in moss, indistinguishable from that used in nest nests with distinctly oblong cup and rim construction, in clumps below the nest. The shapes (Table 1). For all nests (n = 36) the amount found ranged from a few scraps to entire structure ranged from 9.0 to 15.0 cm in that nearly equal to a completed nest. This total height (i.e., bottom of nest to nest rim), extra material, was often a good indication of but often had an additional 4–8 cm of hang- an otherwise hidden nest. ing material below the nest cylinder. Eggs and incubation. Of three nests where laying Nest construction behavior. We observed seven date was determined, one female laid on sub- nests during construction. On all occasions sequent days and two skipped a day between only the female participated. During one eggs. Females spent the night on the nest after observation, a second adult, whose more laying the first egg, but it was not determined glittering frontal patch suggested a male, exactly when incubation began. was seen to swoop repeatedly in and chase All 51 eggs were elliptical and immaculate the female that was building. On three of white. Linear measurements of eggs ranged seven chases this resulted in the loss of the from 14.1–15.7 mm in length and 9.1–10.0 nesting material being carried. At three mm in width (n = 18; see Table 2 for mean ± separate nests, females were observed adding SD). Mean clutch size was 1.92 (n = 26 nests). material to the rim of the nest during incuba- Modal clutch size was two. At 24 nests where tion. clutch size was two, six were nests that were We found one nest as the adult brought in reused, in which a clutch was subsequently the first piece of moss. It appeared that there laid in the same nest, or nests that were rebuilt was already a thick mat of spider web attached in the same location after removal of a suc- to the rock. The moss, an approximately 4-cm cessful nest. One nest, in which two succes- long piece, was hung from this lip. Four days sive clutches were laid, contained only one later a loose, hanging cylinder was formed egg per clutch. At two nests, both with a

325 GREENEY ET AL.

FIG. 3. One to three-day-old Green-fronted Lancebill nestlings begging, January 2002, 2100 m a.s.l., Yanayacu Biological Station, Napo, Ecuador. Photo by H. F. Greeney. clutch of two, the incubation period was 20 especially the gape, became bright yellow to and 21 days, respectively. Only females incu- yellow-orange. As the bill lengthened it dark- bated. Females left the nests as early as 05:48 ened to black, except for the base which h and were never seen to leave the nest after remained yellow. Black pin feathers developed 18:15 h. During the last hours of daylight over the entire body and feather sheaths (17:30–18:15 h), females were often observed broke at ≈6–8 mm in length. Wing pin feath- leaving the nest every 1–2 min to forage for ers broke sheaths around days 8–10. Just prior insects over the stream for periods of 1–4 to fledging, dorsal plumage was green, ventral min. plumage was pale gray, crowns and napes were dull bronzy, and an indistinct white post- Nestlings. Of five nests where hatching dates ocular spot was present. were determined, the eggs hatched one day We examined the crops of two dead, 10- apart in three and on the same day at two. day-old nestlings found in mid-March. The The nestling period was 29 days (two nests crop of one contained 81 small (6–9 mm) with two young each) and 30 days (one nest mayflies (Ephemeroptera) and one 4-mm with a single nestling). long fungus gnat (Sciaridae, Nemotocera, At hatching, nestlings were completely Diptera). The crop of the other nestling con- naked and dark grayish pink with pale creamy tained 74 mayflies, 9 fungus gnats, and one yellow bills. Within 24–48 h the dorsal surface moth fly (Psychodidae, Diptera). turned dark gray to black (Fig. 3). Bills, and The female was almost always observed

326 GREEN-FRONTED LANCEBILL BREEDING BIOLOGY brooding when nestlings were ≤ 10 days old. observed until late September and early Octo- Nests with no adult present contained nest- ber. Nests in all stages were found from Octo- lings that were frequently cold to the touch. ber through December. Most nesting activity Females brooded less frequently once nest- had ceased by the end of January. Adult pres- lings were ≥ 10 days old, and nestlings were ence along streams was greatly reduced by this always warm to the touch. These observations time and by mid-February only an occasional suggest that nestlings begin thermoregulation adult was encountered. A single exception, a sometime around 10 days of age. For addi- nest that had been rebuilt after successfully tional photographs of nests, eggs, and nest- fledging two young and then having been lings, see Sheldon & Greeney (2005). removed, was found with a third clutch of two eggs in mid-February. The nestlings were Nesting success. Seven of 17 nests with two-egg found dead in the nest on 18 March and it is clutches hatched only one egg. Of 19 nests estimated they would have fledged in early to where nest fate was determined, two failed mid-April. We thus estimate that breeding during incubation, eight failed during the activity began in late August to early Septem- nestling period and one failed during building. ber and continued to late March or early April Eight of 19 nests fledged at least one young with the majority of breeding occurring from (42.1% nest-success), with five of these fledg- mid-October to mid-December. ing one young and three fledging two young Three nests were collected after successful (0.58 young per nest, or 1.38 young per suc- fledging. At each of these sites, a second cessful nest). active nest was later discovered in the same Of two nests that failed during incubation, location. At an additional four nests, a second one was destroyed by a waterfall as it clutch was laid after fledging of the first expanded during a rain storm, and a second brood. It is not known whether old nests were collapsed under the weight of the incubating removed and rebuilt or if they were reused, female. A nest that failed during building fell but the short period (< 1 month) elapsing to the ground near completion, possibly due before the appearance of a new clutch sug- to faulty construction. Each of these nests gests that nests were reused. Thus, seven suc- were among those with a great deal of fallen cessful nest sites were reused; no failed nest material below. Two of the eight nests that sites were reused. failed during the nestling period are thought to have done do as a result of predation of the DISCUSSION female away from the nest. Both nests were intact. In one the nestlings appeared Green-fronted Lancebill nesting activity in unharmed and in the other the nestlings northeastern Ecuador began in late August– appeared partially eaten, but the damage was early September and continued through late likely that of insects or snails. The remaining 6 March–early April. The peak in activity nests were damaged and empty, and are pre- appeared to be from mid-October to mid- sumed to have been destroyed by a predator. December. These breeding dates roughly cor- respond to the dry season and are similar to Nesting seasonality and nest reuse. We discovered those reported for Green-fronted Lancebills the first lancebill nest two days prior to laying in Costa Rica (August–January; Stiles & on 8 September 2001. Subsequently, through- Skutch 1989), Colombia (May–January; Hilty out September, we found nests during build- & Brown 1986, Ramirez & Arias 1994), Peru ing and incubation. Nestlings were not (September; Snow & Gochfeld 1977), and

327 GREENEY ET AL. northwestern Ecuador (June–January; Ruschi thus have no measure of variation. If nest 1961, Greeney & Nunnery 2006). Our data material and nest substrate do vary geograph- on the duration of nest building (12–16 days) ically, this may simply reflect variation in avail- and incubation (20–21 days) in Ecuador are ability of nest materials and substrates. very consistent with the 12 day nest-building Documentation of nest reuse and nest site and 20–21 day incubation periods docu- reuse in Colombia (Ramirez & Arias 1994) are mented at a single Green-fronted Lancebill consistent with our observations, and suggest nest in Colombia (Ramirez & Arias 1994). that nest and nest site reuse are common The nestling period at our study site (29–30 occurrences with Green-fronted Lancebills. days) was shorter than the 36 day nestling The systematic relationships of the Dory- period observed at the Colombian nest, but fera lancebills have long been uncertain. The this difference could be due to human activity genus has traditionally been placed next to the patterns near the Colombian nest, which was monotypic genus Androdon (Tooth-billed located at a human dwelling (Ramirez & Arias Hummingbird, A. aequatorialis) in checklists, 1994). Our data on nestling periods are based apparently due to similarities in bill morphol- on a larger sample size (n = 3) and none of ogy (bill serrations of Androdon notwithstand- our nests were near human activity. Our egg ing), and both genera have been included size data are also based on a much larger sam- within the hermit assemblage, subfamily Pha- ple size than data previously available, which ethornithinae (e.g., Cory 1918, Peters 1945). may explain our slightly smaller average egg Systematic work based on osteological char- dimensions (Table 2). acters (Zusi & Bentz 1982) and molecular Nest form (bulky, pendant cup) and nest techniques, including protein electrophoresis location (dark sheltered situations associated (Gill & Gerwin 1989, Gerwin & Zink 1998), with streams) of Green-fronted Lancebills in DNA hybridization (Sibley & Ahlquist 1990, our study area were similar to those reported Bleiweiss et al. 1994, 1997), and DNA elsewhere (Snow & Gochfeld 1977, Hilty & sequencing (Altshuler et al. 2004), however, Brown 1986, Stiles & Skutch 1989). Nest consistently show that Doryfera and Androdon material and attachment to the substrate, are not hermits. Based on similarities in song however, show some variation. Snow and structure, shape and coloration of rectrices, Gochfeld (1977) reported a nest resting on a and hindneck musculature, Schuchmann rock ledge, and Ruschi (1961) described a nest (1995, 1999) suggested that the two genera partially supported by and partially suspended may even constitute a separate subfamily, from a low branch. We also found that, while Doryferinae. On the other hand, recent most nests were completely pendant, some molecular phylogenies indicate that Doryfera received partial support from the substrate and Androdon both belong in the primitive (see Fig. 2). Although Stiles and Skutch (1989) “mango” clade of the Trochilinae (Bleiweiss et reported Costa Rican nests to be usually al. 1997, Altshuler et al. 2004). Among the attached to a rootlet or twig and to include mango-types, which also include Heliactin, tree fern scales, nests in our study area were Heliothryx, Antracothorax and Eulampis, Dory- made exclusively of moss and spider webs fera appears to be most closely related to the and were almost always suspended, attached Colibri violet-ears and the Augastes visor-bear- by the nest rim, from a solid substrate such as ers (Bleiweiss et al. 1997, Altshuler et al. 2004; a rock. It is difficult to assess the importance see also Gerwin & Zink 1998). of this apparent variation because previous The pendant form of Green-fronted Lan- reports were based on very few nests, and cebill nests is similar to that of the phaethor-

328 GREEN-FRONTED LANCEBILL BREEDING BIOLOGY nithine hermits, and is unique within the tion and nestling periods of the Green- Trochilinae. Yet, Green-fronted Lancebill fronted Lancebill are intermediate between nests also show some affinities to the sturdier, those of species thought to be closely related bottom-affixed nests typical of the majority of to lancebills and species that inhabit extremely trochiline species. That is, Green-fronted Lan- high-elevations (e.g., Andean Hillstar, Oreotro- cebill nests are often not completely pendant, chilus estella) (Schuchmann 1999). However, but may receive partial support from the sub- consistent with observations that inclement strate (see also Ruschi 1961) or a ledge of the (i.e., cool, wet) weather often prolongs nest- substrate (see Fig. 2) to which they are affixed. ling development time in hummingbirds (see Thus, while lancebill nests are clearly pendent, Skutch 1973), the relatively long incubation the degree to which they receive support from and nestling periods of the Green-fronted below varies and appears to be intermediate Lancebill may simply be due to the environ- between the truly pendent nests of hermits mental conditions characteristic of its habitat and the typically non-pendent nests of non- and microhabitat. hermits. Our data, therefore, further supports Green-fronted Lancebill nestlings lacked the idea that nest form in Doryfera lancebills the dorsal neossoptiles (natal down) charac- represents a transition between subfamilies teristic of the vast majority of hummingbird Phaethornithinae and Trochilinae. Note that, species (Collins 1978, Schuchmann 1989). while the nest of the single Androdon species, Schuchmann (1989) proposed that dorsal to which Doryfera is traditionally allied, neossoptiles function to stimulate begging remains undescribed, species in all other gen- behavior in older nestlings, and represent an era of the mango clade (following Altshuler et adaptive response to nest predation pressure, al. 2004) construct small compact cups that replacing begging vocalizations. He argued are saddled on horizontal branches or that air currents from an adult female hum- attached to vertical branches (see Schuch- mingbird’s wings, created as she hovers near mann 1999). the nest prior to feeding, are perceived by the Development periods for Green-fronted neossoptiles, which elicits gaping (begging) Lancebills are remarkably long compared with behavior. Schuchmann (1989) showed experi- those documented to date for hermits and mentally that nestlings with neossoptiles most other trochilines (Schuchmann 1999). removed ceased to respond by gaping to a The incubation period for Green-fronted female’s arrival near the nest. One prediction Lancebills was 20–21 days, which is substan- of Schuchmann’s hypothesis is that nestling tially longer than the 16–19 day incubation begging vocalizations should be more preva- period typical for most hummingbirds lent in hummingbird species that lack dorsal (Schuchmann 1999). Among other mango- neossoptiles. In Ecuador, we did not hear types (following Altshuler et al. 2004), incuba- nestling begging vocalizations at Green- tion periods range from 15–18 days in Anthra- fronted Lancebill nests at any stage of nestling cothorax, Augastes, and Colibri, to 17–19 days in development. We note, however, that stream Eulampis (Schuchmann 1999). The nestling noise was characteristic of all Green-fronted period for Green-fronted Lancebills was 29– Lancebill nests that we observed and we can 30 days, also notably longer than the 23–26 not say with absolute certainty that nestlings day nestling periods typical of other hum- did not give begging vocalizations. We mingbirds and, more specifically, the 17–25 encourage field biologists in the Neotropics day nestling periods documented for other to carefully note the presence (and configura- mango-types (Schuchmann 1999). Incuba- tion of) or the absence of dorsal neossoptiles

329 GREENEY ET AL. and begging vocalizations in hummingbird Gill, F. B., & J. A. Gerwin. 1989. Protein relation- nestlings. ships among hermit hummingbirds. Proc. Acad. Nat. Sci. Phil. 141: 409–421. ACKNOWLEDGMENTS Greeney, H. F., & T. Nunnery. 2006. Notes on the breeding of north-west Ecuadorian . Bull. We are indebted to C. Bustamante and M. Br. Ornithol. Club 126: 38–45. Lysinger for allowing access to the San Isidro Hilty, S. L., & W. L. Brown. 1986. A guide to the birds of Colombia. Princeton Univ. Press, Prin- preserve. For logistical support we thank the ceton, New Jersey. owners of the Magic Roundabout Lodge and Peters, J. L. 1945. Check-list of birds of the world. Cabañas, Baeza. We thank C. Torres S. for Volume 5. Harvard Univ. Press, Cambridge, help with Spanish translation. For critically Massachusetts. reviewing earlier versions of this manuscript Ramirez G., M. G., & J. C. Arias G. 1994. Activi- we thank P. R. Martin and M. Lysinger. This dad reproductivo del colibri (Doryfera ludovicae) study was supported by the generous dona- en la Reserva Natural La Planada. Bol. SAO 5: tions of Ruth Ann and John V. Moore 8–12. through the Population Biology Foundation. Ridgely, R. S., & P. J. Greenfield. 2001. The birds of Additional funding was provided by a Pamela Ecuador. Cornell Univ. Press, Ithaca, New & Alexander F. Skutch Award and a Rufford York. Ruschi, A. 1961. Algumas observações sobre: Phae- Small Grant. Preparation of the manuscript thornis yaruqui yarqui (Bourcier), Boissonneaua jar- was greatly facilitated by the generosity of dinii (Bourcier), Doryfera ludoviciae ludoviciae Tim Metz, Jay Peltz, and the Humbolt crew. (Bourcier & Mulsant), Heliangelus viola (Gould), This work was inspired and aided by the Colibri coruscans coruscans (Gould), Helianthea PBNHS and is publication number 16 of the torquata fulgidugula (Gould), Aglaiocercus kingii Yanayacu Natural History Research Group. mocoa (DeLattre & Bourcier) e Aglaiocercus kingii margarethae (Heine). Bol. Mus. Biol. Mello REFERENCES Leitão, Ser. Biol. [Zool.] 27: 1–21. Schuchmann, K. L. 1989. Release of gaping in Altshuler, D. L., R. Dudley, & J. A. McGuire. 2004. hummingbirds (Trochilidae). Wilson Bull. 101: Resolution of a paradox: hummingbird flight at 477–481. high elevation does not come without a cost. Schuchmann, K. L. 1995. Taxonomy and biology Proc. Nat. Acad. Sci. USA 101: 17731–17736. of the Tooth-billed Hummingbird Androdon Bleiweiss, R., J. A. W. Kirsch, & J. C. Matheus. aequatorialis. Mitt. Zool. Mus. Berl. 71 (suppl. 1994. DNA-DNA hybridization evidence for Ann. Orn. 16): 109–113. subfamily structure among hummingbirds. Auk Schuchmann, K. L. 1999. Family Trochilidae 111: 8–19. (Hummingbirds). Pp. 468–680 in del Hoyo, J., Bleiweiss, R., J. A. W. Kirsch, & J. C. Matheus. A. Elliott, & J. Sargatal (eds). Handbook of the 1997. DNA hybridization evidence for the birds of the world. Volume 5: Barn-owls to principal lineages of hummingbirds (Aves: Tro- hummingbirds. Lynx Edicions, Barcelona, chilidae). Mol. Biol. Evol. 14: 325–343. Spain. Collins, C. T. 1978. The natal pterylosis of hum- Sheldon, K. S., & H. F. Greeney. 2005. Doryfera mingbirds. Bull. South. Calif. Acad. Sci. 77: 14– ludovicae, Green-fronted Lancebill. In Greeney, 21. H. F., R. C. Dobbs, & P. R. Martin (eds.). Natu- Cory, C. B. 1918. Catalogue of birds of the Ameri- ral history of Ecuador's mainland avifauna. cas. Field Mus. Nat. Hist. Publ., Zool. Ser, Vol. Available on: http://depts.washington.edu/nhrg/ 13, Pt. 2, No. 1. nhema.html. (Accessed 15 October 2005). Gerwin, J. A., & R. M. Zink. 1998. Phylogenetic Sibley, C. G., & J. E. Ahlquist. 1990. Phylogeny and patterns in the Trochilidae. Auk 115: 105–118. classification of birds: a study in molecular evo-

330 GREEN-FRONTED LANCEBILL BREEDING BIOLOGY

lution. Yale Univ. Press, New Haven, Connecti- johannae). Pp. 549–550 in del Hoyo, J., A. Elliott, cut. and J. Sargatal (eds). Handbook of the birds of Skutch, A. F. 1973. The life of the hummingbird. the world. Volume 5: Barn-owls to humming- Crown Publ. Inc., New York, New York. birds. Lynx Edicions, Barcelona, Spain. Snow, B. K., & M. Gochfeld. 1977. Field notes on Stiles, F. G., & A. F. Skutch. 1989. A guide to the the nests of the Green-fronted Lancebill Dory- birds of Costa Rica. Cornell Univ. Press, Ithaca, fera ludoviciae and the Blue-fronted Lancebill New York. Doryfera johannae. Bull. Br. Ornithol. Club 97: Stotz, D. F., J. W. Fitzpatrick, T. A. Parker, III, & D. 121–125. K. Moskovits. 1996. Neotropical birds: ecology Snow, D. W., & B. K. Snow. 1980. Relationships and conservation. Univ. of Chicago Press, Chi- between hummingbirds and flowers in the cago, Illinois. Andes of Colombia. Bull. Br. Mus. Nat. His. Zusi, R. L., & G. D. Bentz. 1982. Variation of a Zool. 38: 105–139. muscle in hummingbirds and swifts and its sys- Stiles, F. G. 1999. Green-fronted Lancebill (Doryfera tematic implications. Proc. Biol. Soc. Wash. 95: ludovicae) and Blue-fronted Lancebill (Doryfera 412–420.

331